Fault detection and isolation for gyro and infrared earth sensor based on observer and filter

Abstract

The gyro and the infrared earth sensor (IES) are two kinds of important components in satellite attitude control system. A fault detection and isolation (FDI) approach based on observer and Kalman filter is presented. Firstly, the models of gyro and IES are discussed, and a virtual linear system of satellite pitch channel is constructed. Then, an observer for fault detection and a Kalman filter for fault isolation are designed. The observer could detect faults from IES and gyro, but the Kalman filter could only detect IES fault rather than sensitive to gyro fault. So, fault isolation method could be achieved via fault arbitration logic given in the paper. Lastly, computer simulations are finished on the condition of system health status (no faults), IES fault and gyro fault. The simulation results show that the proposed fault detection and isolation approach is correct and feasible.